[
https://issues.apache.org/jira/browse/IGNITE-10721?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16726232#comment-16726232
]
Alexey Kosenchuk commented on IGNITE-10721:
-------------------------------------------
[~isapego]
In my opinion, it's not clear what is "64 most significant bits of UUID" and
"64 least significant bits of UUID". Because what is UUID is not clarified.
As I wrote above, if UUID here is https://tools.ietf.org/html/rfc4122 then it
should be specified how it's fields (time_low, time_mid,... and all other) are
binary encoded here.
Or UUID here is any generic 128-bit number, not necessary RFC-4122? That should
be clearly explained, I believe.
In any case, an example of binary encoding may help.
Also, why the current 2.7 spec is updated?
I think, such update of the spec should go into the new version.
> Documentation: Fix UUID thin client format description
> ------------------------------------------------------
>
> Key: IGNITE-10721
> URL: https://issues.apache.org/jira/browse/IGNITE-10721
> Project: Ignite
> Issue Type: Task
> Components: documentation, thin client
> Reporter: Igor Sapego
> Assignee: Igor Sapego
> Priority: Major
> Fix For: 2.8
>
>
> UUID thin client format description [1] need to be fixed. The actual format
> of the UUID should be two longs, not a single 128-bit value. Two longs
> written in little-endian are not equal to one 128-bit number.
> [1] -
> https://apacheignite.readme.io/docs/binary-client-protocol-data-format#section-uuid-guid-
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
